RESEARCH Open Access

Managing daily intensive care activities: Anobservational study concerning ad hoc decisionmaking of charge nurses and intensivistsHeljä Lundgrén-Laine1,2*, Elina Kontio1,3, Juha Perttilä4, Heikki Korvenranta2, Jari Forsström5 and Sanna Salanterä1,2

Abstract

Introduction: Management of daily activities in ICUs is challenging. ICU shift leaders, charge nurses and intensivistshave to make several immediate ad hoc decisions to enable the fluent flow of ICU activities. Even though themanagement of ICU activities is quite well delineated by international consensus guidelines, we know only a littleabout the content of the real clinical decision making of ICU shift leaders.

Methods: We conducted an observational study with the think-aloud technique to describe the ad hoc decisionmaking of ICU shift leaders. The study was performed in two university-affiliated hospital ICUs. Twelve chargenurses and eight intensivists were recruited. Observations were recorded and transcribed for qualitative contentanalysis using the protocol analysis method. The software program NVivo 7 was used to manage the data. Theinterrater agreement was assessed with percentages and by Cohen’s �.

Results: We identified 463 ad hoc decisions made by the charge nurses and 444 made by the intensivists. Duringour data collection time, this breaks down to over 230 immediately made decisions per day (24 hours). We dividedthe ad hoc decision making of ICU shift leaders into two types: process-focused and situation-focused. Process-focused decision making included more permanent information, such as human resources, know-how and materialresources, whereas situation-focused decision making included decisions about single events, such as patientadmission. We named eight different categories for ICU ad hoc decision making: (1) adverse events, (2) diagnostics,(3) human resources and know-how, (4) material resources, (5) patient admission, (6) patient discharge, (7) patientinformation and vital signs and (8) special treatments.

Conclusions: ICU shift leaders make a great number of complex ad hoc decisions throughout the day. Often thisdecision making involves both intensivists and charge nurses. It forms a bundle that requires versatile, immediateinformation for a successful outcome. In the future, we need to investigate which information is crucial for ad hocdecision making. These challenges should also be emphasised when information technology programs for ICU caremanagement are developed.

IntroductionMany and varied multiprofessional decisions are madewhen running daily ICU activities. These result in bothclinical and managerial orders to support patient careand ICU work flow. Decision making related to immedi-ate care must often be made quickly in response to thechanging condition of the patient. This decision makingis usually supported by patient information systems that

are designed to provide detailed individual, patient-focused information [1]. However, decision making con-cerning direct patient care is difficult or even impossibleto complete if the managerial decisions are made inac-curately and/or with a delay.Within fast-paced ICU environments, many of these

managerial decisions have to be made immediately, adhoc. By ‘ad hoc decision making’, we mean critical jud-gements that are needed for a specific purpose at a pre-cise moment with the goal of ensuring instant andadequate patient care and a fluent flow of activities inthe ICU. Shift leaders, intensivists and charge nurses in

* Correspondence: hklula@utu.fi1Department of Nursing Science, University of Turku, Lemminkäisenkatu 1,FI-20014 Turku, FinlandFull list of author information is available at the end of the article

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

© 2011 Lundgrén-Laine et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.

ICUs are responsible for coordinating, planning andassessing the daily activities of the units and are thosewho usually make these managerial ad hoc decisions.ICU shift leaders’ duties and tasks are a mixture ofdirect patient care, supervision of care and administra-tive work. This is why many decisions have to be con-sidered both from the clinical and the managerial pointsof view.The decision making of ICU professionals has been

the focus of several studies (see, for example, [2-5]). Inaddition, the need for teamwork, verbal communicationand information has been found to be crucial for taskcoordination and work performance in ICUs [6-8].However, the focus of previous studies has been mainlyon the individual level or on direct patient care.In recent years, precise guidelines and best practice

instructions have been developed to guide the work flowof ICUs and managerial decision making. Despite theseefforts, the ad hoc decision making of ICU shift leadersis poorly supported by information systems. It is com-mon for information needed for decision making to begathered into different information systems and otherplatforms. In addition, part of the decision making isbased solely on the memory and work experience of theICU shift leader. With a growing information load, it isessential that precise, consistent and timely informationbe available for accurate and correct decision makingexactly when it is needed [9,10]. However, we do nothave a clear picture of what decisions the ICU shift lea-ders make.In this study, we focused on the unit-level decisions

which are made ad hoc by ICU shift leaders to ensurethe fluent flow of activities. The specific purpose of thestudy was to identify the ad hoc decisions of ICU shiftleaders by evaluating their cognitive processes duringtheir daily work.

Materials and methodsContext and settingWe conducted this study in two university hospital ICUsin Finland from 25 April to 23 June 2007. Both of theICUs are mixed medical-surgical units with 24 (ICU1)and 22 (ICU2) beds Both units are run by full-timeintensivists with 24-hour coverage. Together the unitstake care of over 4,000 patients annually. In-houseintensivists and qualified charge nurses manage the dailyactivities of the units in both ICUs. The ICUs areclosed-model units in which patients are admitted bythe ICU physician who specialises in surgery, anaesthe-sia or internal disease. The charge nurses of the unitsare registered nurses experienced in directing nursingstaff. Decisions related to direct patient care are madejointly by the specialist consultants and the ICU team.Decisions concerning ICU work flow are made within

the ICU team. Both of the study units use Centricity™Critical Care [11] information systems for electronicpatient documentation. More detailed characteristics ofthe ICUs are given in the Table 1.

Study designWe performed an observational study using the think-aloud technique in a real clinical setting. The basis ofthe think-aloud technique is to ask participants to ver-balise their performance while performing their tasksand duties. Researchers have found this technique to bean appropriate method for investigating and revealingcomplex cognitive processes in a real-life context[12,13]. Our purposeful sample consisted of 12 chargenurses and 8 intensivists equally at both hospitals. Weconducted the collection of data during different shifts:15 morning shifts, 4 evening shifts and 1 night shift.Data collection was also performed on different days ofthe week, including three weekend shifts. The data col-lector followed each participant individually from thebeginning of the eight-hour work shift. The role of thedata collector was to remind participants to talk aloudduring the observations. All sessions were recordedusing an MP3 player. One researcher (HLL), who wasfamiliar with intensive care, conducted all the sessions.

Data analysisThe data were typed out into a .txt file in an authenticform. Since the recorded data included each partici-pant’s speech during his or her shift, irrelevant partsthat were not related to the focus of our study’s purposewere omitted. Before the analyses, the text files wereread several times and situation-related notes weremade. These situation-related notes were made to detectwhich decision was related to which situation at thetime of observations (for example, staff ratios to isola-tion). This supported the final analyses and

Table 1 Characteristics of the participating ICUs

Characteristics, 2007 ICU 1 ICU 2

Patients, n 1,727 2,615

Surgery 71.4% 56.3%

Conservative treatment 28.6% 43.7%

Mean LOSa, days 3.4 2.0

Mortality 7.8% 3.8%

Fellows, n 4 4

Residents, n 1 1 or 2

Assistant physicians, n 3 N/A

Head nurses, n 2 1

Staff nurses, n 3 2

Registered nurses, n 112 69

Practical nurses, n 9 12aLOS = length of stay; N/A = not applicable.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 2 of 10

understanding of the rapidly changing situations andICU-specific concepts related to the complex thinkingof the study participants. Qualitative content analysiswas then used to identify each decision made by theparticipants. Qualitative content analysis is ‘a researchmethod for subjective interpretation of the content oftext data through a systematic classification process ofcoding and identifying themes or patterns’ [14] (p.1278). Each recording was analysed separately with theaim of detecting the decisions of both professionalgroups.Our analysis consisted of three phases: free-model

coding, tree-model coding and application of the proto-col analysis. A computer-assisted qualitative softwareprogram, NVivo 7 (QSR International, Doncaster, Vic-toria, Australia) [15], was used to support the manage-ment of the coding in our analysis. First, the free-modelcoding was used to identify decisions made by the parti-cipants. This meant that all utterances which includeddecision making were coded. Then we continued toorganize the decisions hierarchically into clusters (tree-model coding). Some of the hierarchical single codingswere still in several categories during this phase.In the next phase, the hierarchical categories were

compared and those with close contents were inte-grated, such as human resources and the know-how ofpersonnel. All the coding phases were first performedby one researcher (HLL). After this, two researchers(HLL and EK) independently analysed all of the pre-coded data. We ensured the reliability with intercoderagreement percentages and the consistency of codingwith Cohen’s � values (Table 2).Finally, we applied protocol analysis and analysed the

levels of verbalization in sentences. This was done toidentify ad hoc decisions, that is immediately made

decisions. In the protocol analysis, the verbalization ofstudy subjects was classified into three levels. Theselevels were used to describe the reproduction of infor-mation from the memory of a study subject while he orshe was performing a task. First- and second-level ver-balizations were considered reliable because these levelsare assumed to reveal the content of the short-termworking memory [12]. By applying protocol analysis, wewere able to code each decision into a different level.First- and second-level decisions were considered adhoc decisions. Verbalizations of decisions at levels 1 and2 were clearly and immediately expressed as well asclearly targeted. These decisions needed to be maderapidly, and the decision maker sought either animmediate solution or more information to be able toconclude his or her decision without delay. For thethird-level decisions, participants needed additionalinformation and the solutions required for decisionmaking were not immediately essential. The third-leveldecisions were not ad hoc decisions and included thoserelated to a patient discharge plan, such as, ‘We aregoing to extubate probably tomorrow, and if all goesokay he will be discharged’.A study on protocol analysis by Ericsson and Simon

[12] and an article by Lundgrén-Laine and Salanterä[16] discussed the methodological bases of protocol ana-lysis and the levels of the analysis in more detail.

Ethical issues and participantsThe study was approved by the hospital districts’ autho-rities. In our study, we followed the Finnish national leg-islation and ethical principles [17]. No data arepersonally identifiable. The inclusion criteria for thestudy subjects were that the participant had to (1) beable to voluntarily, capably and competently talk aloud

Table 2 Categories, amounts, coding frequencies of ad hoc decisions, number of ad hoc decisions, interrater reliabilityvalues and observation times

Categories of ad hoc decisions Shift leaders Intensivists (n = 8) Charge nurses (n = 12)

Process-focused, n (%)

1. Human resources and know-how 291 (32.1%)b 291 (63%)b

2. Material resources 22 (2.4%) 22 (4.8%)

Situation-focused, n (%)

3. Patient admissions 32 (3.5%) 3 (0.7%) 29 (6%)

4. Patient information and vital signs 246 (27%)b 174 (39%)b 72 (16%)b

5. Special treatments 171 (19%)b 147 (33%)b 24 (5%)

6. Diagnostics 86 (9.5%) 86 (19%)b

7. Adverse events 1 (0.1%) 1 (0.2%)

8. Patient discharges 58 (6.4%) 34 (8%) 24 (5%)

IRRa (%)/Cohen’s � 97.0/0.92 to 1.0 91.0/0.90 to 1.0

Total ad hoc decisions, n 907 444 463

Total observation time, hours 92 30 62aIRR = interrater reliability; blargest categories.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 3 of 10

during work situations and (2) have current ICU shiftleadership experience. A written consent form wasobtained from each participant, and each received writ-ten and oral instructions before the observations.

ResultsFour of the intensivists were men and four were women,with their ICU work experience ranging from 4 to 22years. Three of the twelve charge nurses were men, andnine were women. Their ICU work experience rangedfrom 5 to 32 years. None of the participants withdrewfrom our study during the observations.The recordings varied from two to six hours per infor-

mant (mean duration 4 hours, 40 minutes), and thelength of the recordings depended on the intensity ofthe shift. Altogether we recorded 92 hours of data. Inour analysis, we identified 444 ad hoc decisions made bythe intensivists and 463 ad hoc decisions made by thecharge nurses. This breaks down to nearly 10 ad hocdecisions per hour or 240 ad hoc decisions per daymade by the ICU shift leaders. If we consider the num-ber of decisions per decision maker, these data breakdown to almost 56 ad hoc decisions per intensivist and39 decisions per charge nurse.The ad hoc decision making of intensivists and charge

nurses covered the entire patient care process fromadmission to discharge. On the basis of our analysis, weidentified eight decision-making categories: (1) adverseevents, (2) diagnostics, (3) human resources and know-how, (4) material resources, (5) patient admission, (6)patient discharge, (7) patient information and vital signsand (8) special treatments.These categories were in turn divided into two differ-

ent types of decision making: process-focused and situa-tion-focused (Figure 1). In our data, the process-focuseddecision making represented ad hoc decision makingthat was related to permanent events. In addition, theprocess-focused ad hoc decisions were made concerningthe entire unit, and they affected the work flow of thewhole ICU. Human resources, know-how and materialresources represent process-focused decision making(Figure 1, horizontal type).The situation-focused ad hoc decision making con-

cerned incidents that occurred at a certain moment.Typically they were single incidents or concerned onepatient or individual, such as patient admission, adverseevents, diagnostics, patient information and vital signs,special treatments and patient discharge (Figure 1, verti-cal type).Most ad hoc decisions belonged to the categories of

human resources and know-how (32.1%), patient infor-mation and vital signs (27%) and special treatments(19%), representing both process-focused and situation-focused decision making. These categories were followed

by diagnostics (9.5%) and patient discharge (6.4%). Thefewest ad hoc decisions made were related to patientadmission (3.5%), material resources (2.4%) and adverseevents (0.1%).Table 2 summarises the named ad hoc decision-mak-

ing categories, the frequencies and percentages of ourcoding, the participants in the study, the total observa-tion times and intercoder reliability values with Cohen’s� values. In Table 3, we present the definitions for thenamed ad hoc decision-making categories.

Human resources and know-howMost of the ad hoc decisions made by ICU shift leaderswere about human resources and know-how (32.1%).The main objective of this decision making was to man-age the number of ICU personnel, to ensure sufficientand appropriate resources or to compensate for theknow-how levels needed so that patient care could beensured in all situations around the clock. Remarkably,all the ad hoc decisions in this category were made bythe charge nurses; however, they typically made thesedecisions in coordination with other ICU colleagues ornegotiated in multidisciplinary teams before personallymaking judgments. In addition, many of the ad hocdecisions identified in this study dealt with a differentkind of follow-up data needed for statistics or research.Typical ad hoc decisions under this category are pre-sented in Table 3.

Patient information and vital signsThe second most frequent ad hoc decisions were madeconcerning patient information and vital signs (27%).This category covered very intensive care, specificpatient information and the present condition of apatient fulfilling the criteria for essential ICU care.These ad hoc decisions dealt, for example, with issuesrelated to the mode of isolation, patient care intensity,critical problems regarding vital functions or changesrequired immediately for the intensive care plan. The adhoc decision making in this category was also closelyconnected to other ad hoc decisions, such as those con-cerning admissions or discharges. For example, highpatient care intensity led to the next ad hoc decision:which of the patients would stay and who would bemoved to another ward. About 40% of these decisionswere made by the intensivists, and these decisions com-prised most of the ad hoc decisions they made (Table2). The ad hoc decisions under this category are pre-sented in Table 3.

Special treatmentsAd hoc decision making related to special treatments(19%) was the third-largest category. Under this cate-gory, we named all those ad hoc decisions which dealt

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 4 of 10

with ICU-specific care and administration of medication.ICU-specific care covered the treatments which werepossible to perform only under intensive care circum-stances, such as hyperbaric oxygen therapy, care ofpatients with intra-aortic balloon pumps or treatmentsperformed by specialised consultants. Under this cate-gory, we also named administration of ICU-specificmedication, such as medication that requires intensivemonitoring (for example inotropes, vasodilators or seda-tives). Examples of these ad hoc decisions are presentedin Table 3.

Diagnostics and patient dischargePart of the ad hoc decision making of ICU shift leaderswas related to diagnostics and patient discharge (9.5%and 6.4%, respectively). Only the intensivists made adhoc decisions about diagnostics, covering one-fifth of allad hoc decisions they made. An example of an ad hocdecision related to diagnostics is, ‘Diagnosis confirmedtoday induces immediate changes in the patient’s careplan’. The ad hoc decisions concerning patient dis-charge, such as, ‘This patient is ready for discharge’,were made by both the intensivists and the chargenurses (see also Table 3).

Patient admission, material resources and adverse eventsAd hoc decisions about patient admission, materialresources and adverse events constituted only a smallportion of decisions made by the shift leaders. Intensi-vists and charge nurses made decisions about patientadmission. Examples of these ad hoc decisions are ‘Fourbeds should be booked for elective patients during thepresent shift’ or ‘The patient from the ER should beadmitted immediately’. Decisions about materialresources and adverse events were all identified fromthe data sets of the charge nurses. Examples of these are‘A special mattress is needed for the incoming patient’or ‘The pinprick accident should be managed immedi-ately according to the treatment protocol’ (see alsoTable 3).

A bundle of ad hoc decisionsMost of the ad hoc decisions made by the ICU shift lea-ders were not isolated decisions but event-based, com-plex combinations of several decisions made by both theintensivist and the charge nurse. In addition, the ad hocdecision making of ICU shift leaders did not follow alinear process, and it varied between decision makers aswell as in changing situations. The ad hoc decision

AD HOC 1 Admission

(see in detail Figure 2)

AD HOC 2 Admission

(see in detail Figure 2)

AD HOC 3 Admission

(see in detail Figure 2)

SITUATION-FOCUSED

PROCESS-FOCUSED

Patie

nt a

dmis

sion

Patie

nt in

form

atio

n an

d vi

tal s

igns

Spec

ial t

reat

men

ts in

clud

ing

med

icat

ion

Dia

gnos

tics

incl

udin

g la

bora

tory

and

imag

ing

Adve

rse

even

ts

Patie

ntdi

scha

rge

Human resources and know-how

Material resources

Figure 1 Process-focused (horizontal) and situation-focused (vertical) ad hoc decision making.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 5 of 10

making was a multidisciplinary process in which theroles of ICU shift leaders were not so well-defined. Onead hoc decision made by either of the shift leaders cre-ated a bundle of ad hoc decisions that were needed tocomplete the task. Typically, both process-focused andsituation-focused ad hoc decision making was thenneeded. Figure 1 and ad hoc decisions 1, 2 and 3,explained in detail in Figure 2, show some simplifiedexamples of the ad hoc decision-making bundles duringthe daily management of ICU activities. These exampleshave been extracted from our data. For example, differ-ent kinds of ad hoc decisions have to be made eachtime a new patient is admitted to the ICU, dependingon various factors such as patient condition, urgencylevel, staffing and patient load.

DiscussionTo our knowledge, our study is the first to describe thead hoc decision making of ICU shift leaders, includingintensivists and charge nurses, during the managementof daily ICU activities. Our study shows that ICU shiftleaders make numerous ad hoc decisions when mana-ging the daily activities of their units. Our analysisrevealed that almost 10 immediately needed ad hocdecisions per hour are made by the shift leaders, which

presents great challenges for immediate informationretrieval.We identified eight ad hoc decision-making categories

for ICU shift leaders: (1) adverse events, (2) diagnostics,(3) human resources and know-how, (4) materialresources, (5) patient admission, (6) patient discharge,(7) patient information and vital signs and (8) specialtreatments. These categories covered the whole ICUcare process from patient admission to patient dis-charge. Even if the ad hoc decisions of the shift leadersdiffered, the goals and the content underlying their deci-sions were parallel. For example, intensivists made adhoc decisions about incoming patients and who wouldbe admitted first (patient admission). At the same time,charge nurses, for their part, made ad hoc decisionsabout the timing and placement of incoming patients(patient admission).Most of the identified ad hoc decisions of the ICU

shift leaders were connected to intensive care-specificpatient information and the condition of the patient,human resources and know-how of the personnel andspecial treatments performed in intensive care settings(nearly 80%). This result confirms the basic role of ICUwork, in which the most critically ill patients are takencare of by multiprofessional teams and the most

Table 3 Definitions of ad hoc decision-making categories and examples

Ad hoc category Definition Examples

1. Adverse events An injury related to medical management ’Concerning this pinprick accident, I have to order ablood test for her immediately’.

2. Diagnostics Patient diagnosis, including laboratory and radiology resultsaffecting ICU work organization

’Today’s X-ray is OK, so she will be ready to leave’.

3. Human resourcesand know-how

The amount of staff resources, the knowledge of the ICUpersonnel and follow-up information and reports

(1) ‘In that room, there is one very experienced nurse. Itis enough. I’ll keep it that way’.

(2) ‘Two RNs are needed for this patient in the followingshift’.

(3) ‘RN XX will be named in the trauma team in thefollowing shift’.

4. Material resources Materials needed for ICU patient care ’This isolation room is reserved for this patient in the nextshift’.

5. Patient admission Acceptance of patients for ICU care ’The recovery room is booked up, so we have to take thispatient immediately’.

6. Patient discharge Acceptance of patient transfer from ICU care ’If the bleeding stops, he will be discharged when a newpatient is introduced’.

7. Patient informationand vital signs

Intensive care-specific patient information and patient’s intensivecare- specific condition

(1) ‘We will put her to sleep and then deal with the AF’.a

(2) ‘The patient with the highest nursing intensity willneed two nurses in the evening shift’.

(3) The patient’s invasive blood pressure values are toolow, and he is not ready for discharge’.

8. Special treatments ICU-specific care and medication administration (1) ‘We will make the sterna closure tomorrow, socardiologist consultation today’.

(2) ‘Our unit is ready to admit a patient who needshaemodialysis’.

(3) ‘With this patient, we will start an inotrope infusionand ask for a cardiology consultation’.

aAF = atrial fibrillation.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 6 of 10

demanding treatments can be performed by an adequatenumber of team members and specialised staff. In addi-tion, our study results suggest that the ad hoc decisionmaking of ICU shift leaders is complex and overlapping.These findings are supported by previous research onthe decision making of ICU personnel [18,19].Most ad hoc decision making concerns human

resources and know-how. Similar results have beenfound in previous studies [7,20] in which the informa-tion requirements of the ICU patient care team andmultidisciplinary care team in an emergency departmenthave been investigated. Results of previous studies haveshown that many clinical questions contain organiza-tional elements which are essential for the coordinationof the work.Both our study and those of Reddy et al. [7,20]

showed that clinical decision making is an importantpart of ad hoc decision making, and it is often con-nected with organizational issues. In our study, the clini-cal ad hoc decisions were mainly made by theintensivists, followed by several ad hoc decisions relatedto organizational issues made by charge nurses. InReddy and Spence’s study [7], for example, the patients’conditions were evaluated in relation to the need forbeds. Similar aspects were found in our data. An ad hocdecision made by the intensivist regarding special treat-ment created several ad hoc decisions related to organi-zational issues such as placement of the patient, the

availability and capability of the nurse, the timing of thetreatment and ensuring the continuum of care fromshift to shift.In our study the charge nurses made most of the

organizational decisions, which mainly concernedhuman resources and know-how. These organizationalelements seem to contribute significantly to the workflow of the ICU. Our results might reflect the organiza-tional culture of the ICUs investigated and how thework was organised and shared by the ICU team. Itwould be interesting to study this issue further in otherICUs at both the national and international levels.The amount of different ad hoc decisions gives us

information about which ad hoc decisions dominate theICU shift leaders’ work. However, it does not reveal theimportance of the ad hoc decisions made. In our analy-sis, we found that the fewest ad hoc decisions madewere related to diagnostics, patient discharge, patientadmission, material resources and adverse events. How-ever, all of these include extremely important decisionswhich affect the outcomes of ICU patient care.Especially in urgent situations, when, for example,

patient admission and discharge decisions have to bemade immediately, all decisions should be based onaccurate, clearly defined information. In previous stu-dies, delays in patient admissions or after-hours dis-charges, for example, have been found to be associatedwith patient mortality [21,22]. Diagnostics play an

AD HOC 3Admission of a

patient with minorblood pressure

problems

Urgent

Placing

Number of nurses

NO

AD HOC 1Admission of apatient from theangiography unit

Resuscitated

Urgent

Placing

Number of nurses

Mechanical ventilation

YES

YES

YES

AD HOC 2Admission of a

patient with acuterenal failures

Special treatmentneeded

Urgent

Placing

Number of nurses

Know-how and skillsContinuity of care inthe next shift

YES

YES

Additional materialsfor the treatment

Figure 2 Examples of ad hoc decision-making bundles.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 7 of 10

important role in patient care in the ICU, and this cate-gory is closely connected to both admission and dis-charge decisions as well as to immediate changes thatare required in care plans. Ad hoc decision makingrelated to adverse events is often vital, since poorlythought out decisions can cause serious complicationsand even patient mortality [23,24].ICU care management, coordination and decision-

making procedures are strongly supported by the inter-national consensus guidelines [25-27]. These guidelinescontain definitions and objectives for the quality of ICUperformance. In addition, recommendations for optimalICU care are provided. However, these guidelines havemainly focused on the material resources needed at cer-tain ICU care levels or in direct patient care, as well ason vital problems of ICU patients. Process-focused deci-sion making has received little attention, and improve-ments in process performance from the multiprofessionalpoint of view are needed [28,29].Our results reveal that studies concerning only one

professional group or one decision-making area are notenough when complex and fast-paced environmentssuch as ICUs are investigated. In our study only veryfew ad hoc decisions were made by one person. Insteadthey were a chain of decisions made by several persons.The ad hoc decision making of ICU shift leaders is ateam process whereby the final outcome is achievedwith the collaboration of charge nurses and intensivists,which requires that several ad hoc decisions be made.Our study has some strengths and some limitations.

One limitation of our study is that we conducted it inonly two ICUs and in one country. However, withinthese two ‘full-service’ ICUs, both of which have largeICU patient populations, we managed to obtain rich andsaturated data. Our study ICUs were not randomlyselected; rather, their selection was purposeful andbased on the facts that both units take care of patientswith various medical and surgical problems and bothhave similar operation models in their care organization.A purposeful sample also offered data of good quality.In addition, we did not evaluate the diurnal variation indecision-making frequency, which would be interestingto assess in the future. We hypothesized that more ICUdaily activities are performed during morning shifts andcreated more stress on those choosing the 15 morningshifts to be observed. However, we suggest that the var-iation in decision-making frequency is more dependenton clinical situations than on working shifts. Our studydoes not provide information about the quality of shiftleaders’ decisions (that is, good and bad decisions). Toidentify the quality of the ad hoc decisions made, thewhole decision-making process should be followed untilthe end and the final outcomes related to each decisionshould then be evaluated. Because of the type of data

we collected, we cannot generalise our results per se,but other researchers can utilise the results in futureresearch by testing them in other ICU settings.A challenge in our study was combining the think-

aloud technique with protocol analysis, especially thedifficulty of compiling research on decision making inreal settings. Think-aloud is used quite often in observa-tional studies in health care [4,30,31]. However, protocolanalysis in combination with think-aloud has rarely beenused in acute clinical settings in health care. Alterna-tively, we could have used pure observation or intro-spection as a method of investigating the ad hocdecision making of shift leaders. However, with observa-tion, a researcher might miss or misread a significantpart of decision making. On the other hand, introspec-tion involves interpretations by the participants them-selves, which might affect the results (see, for example,[32]). In our study, we were interested in evaluating adhoc decisions that had to be made immediately. It is notlikely that the Hawthorne effect would play a significantrole in this kind of decision-making situation. In ourstudy, the think-aloud technique combined with proto-col analysis functioned well. The participants in ourstudy were capable of verbalizing their thinking. In addi-tion, with protocol analysis, we were able to find andclassify the ad hoc decision making.The observational study design itself is a big challenge

when complex phenomena are investigated in ICU set-tings. Even the shadowing of one participant whosework is to make many different decisions all the timewas demanding. In our study, we used only one observerand recorder with one participant during one shift withthe aim of covering the whole ICU situation as well aspossible. Many details should be considered carefullywhen more than one decision maker, or shared decisionmaking, is observed simultaneously. These details are,for example, time labels in starting and ending multipledecisions and differentiating ad hoc decisions aimed atthe same targets from other decisions made at the sametime. Combining handheld computers with work obser-vations as Westbrook and Ampt tested in their study[33] using our study method could be an entirely novelway to look at this phenomenon in the future.In the future, the reliability and generalization of our

results should be validated by national and internationalmulticentre studies in which the information variablesthat are crucial for ad hoc decision making in the ICU.The international perspective is supported by the factsthat the basic targets of ICU care are internationallyquite similar and the decision making of ICU personnelhas also been found to be internationally comparable[34]. In addition, the processes of ICUs are internation-ally quite well defined and guided by international con-sensus guidelines supporting the managerial decision

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 8 of 10

making of ICU professionals and standardizing the poli-cies of the ICUs. However, different cultures and teamfunctions might have some effect on decision-makingprocesses in different ICUs.

ConclusionsOur study describes the ad hoc decision making of ICUcharge nurses and intensivists during the daily manage-ment of ICU activities. This study highlights the com-plex and fundamental phenomenon of ad hoc decisionmaking in a real clinical ICU context. We have shownthat ICU shift leaders make a vast number of ad hocdecisions concerning the entire ICU care process. Orga-nizational issues are greatly emphasised. In addition, thead hoc decision making of ICU shift leaders appears tobe a multiprofessional process in which several ad hocdecisions are needed to complete one task. By identify-ing ICU shift leaders’ ad hoc decisions and this multi-professional process, we will be able to reveal what kindof information is needed in challenging clinical settings.Further research is required to identify what informationneeds are fundamental for ICU shift leaders in ad hocdecision-making situations. In the future, this will helpus to develop electronic decision-making support andmanagement systems for ICUs.

Key messages• ICU shift leaders make numerous immediate decisionsduring the management of daily activities.• The ad hoc decision making during the management

of daily ICU activities is a multiprofessional process, andthe final outcome is usually achieved through multipleimmediate decisions, a bundle of ad hoc decisions.• The ad hoc decision making of ICU shift leaders is

not a linear process, and it involves various areas of ICUcare. The organizational issues are highly emphasisedthroughout the daily management of ICU activities.• In the future, the ad hoc decision making of ICU

shift leaders and access to the most crucial informationshould be confirmed.

AcknowledgementsThe authors gratefully thank the participating units for their collaborationand the charge nurses and intensivists who attended the studyobservations. This study was funded by the Finnish Funding Agency forTechnology and Innovation, Tekes (grant 40020/07), the Finnish CulturalFoundation and The Hospital District of Southwest Finland special grant-in-aid (EVO). Part of this study was presented at the 21st Annual Congress ofthe European Society of Intensive Care Medicine (ESICM) in September 2008in Lisbon, Portugal.

Author details1Department of Nursing Science, University of Turku, Lemminkäisenkatu 1,FI-20014 Turku, Finland. 2Bureau of Administration, Turku University Hospital,PO Box 52, FI-20521 Turku, Finland. 3Faculty of Telecommunication and E-Business, Turku University of Applied Sciences, Joukahaisenkatu 3 A, FI-20520Turku, Finland. 4Department of Anaesthesia and Intensive Care, Turku

University Hospital, PO Box 52, FI-20521 Turku, Finland. 5The Finnish MedicalAssociation (FMA), PO Box 49, FI-00501 Helsinki, Finland.

Authors’ contributionsHLL designed the study, performed the observations, analysed the data anddrafted the manuscript. EK acted as an intercoder and made comments onthe manuscript. HK, JP and JF helped to draft the manuscript. SS supervisedthe study, assisted in the design of the study and prepared the finalmanuscript with HLL. All authors read and approved the final manuscript.

Competing interestsThe authors declare that they have no competing interests.

Received: 13 April 2011 Revised: 11 July 2011 Accepted: 8 August 2011Published: 8 August 2011

References1. Ward NS, Snyder JE, Ross S, Haze D, Levy MM: Comparison of a

commercially available clinical information system with other methodsof measuring critical care outcomes data. J Crit Care 2004, 19:10-15.

2. van Lieshout EJ, de Vos R, Binnekade JM, de Haan R, Schultz MJ, Vroom MB:Decision making in interhospital transport of critically ill patients:national questionnaire survey among critical care physicians. IntensiveCare Med 2008, 34:1269-1273.

3. Fackler JC, Watts C, Grome A, Miller T, Crandall B, Pronovost P: Critical carephysician cognitive task analysis: an exploratory study. Crit Care 2009, 13:R33.

4. Aitken LM, Marshall AP, Elliott R, McKinley S: Critical care nurses’ decisionmaking: sedation assessment and management in intensive care. J ClinNurs 2009, 18:36-45.

5. Garg AX, Adhikari NK, McDonald H, Rosas-Arellano MP, Devereaux PJ,Beyene J, Sam J, Haynes RB: Effects of computerized clinical decisionsupport systems on practitioner performance and patient outcomes: asystematic review. JAMA 2005, 293:1223-1238.

6. Reader TW, Flin R, Mearns K, Cuthbertson BH: Developing a teamperformance framework for the intensive care unit. Crit Care Med 2009,37:1787-1793.

7. Reddy MC, Spence PR: Collaborative information seeking: a field study ofa multidisciplinary patient care team. Inf Process Manag 2008, 44:242-255.

8. Aslakson RA, Wyskiel R, Shaeffer D, Zyra M, Ahuja N, Nelson JE,Pronovost PJ: Surgical intensive care unit clinician estimates of theadequacy of communication regarding patient prognosis. Crit Care 2010,14:R218.

9. Manor-Shulman O, Beyene J, Frndova H, Parshuram CS: Quantifying thevolume of documented clinical information in critical illness. J Crit Care2008, 23:245-250.

10. Najjar-Pellet J, Jonquet O, Jambou P, Fabry J: Quality assessment inintensive care units: proposal for a scoring system in terms of structureand process. Intensive Care Med 2008, 34:278-285.

11. Centricity™ Critical Care. [http://www.gehealthcare.com/euen/iis/products/acute-care/iis_acute_care.html].

12. Ericsson KA, Simon HA: Protocol Analysis: Verbal Reports as Data. Revisededition edition. Cambridge, MA: MIT Press; 1993.

13. Kushniruk AW: Analysis of complex decision-making processes in healthcare: cognitive approaches to health informatics. J Biomed Inform 2001,34:365-376.

14. Hsieh H, Shannon S: Three approaches to qualitative content analysis.Qual Health Res 2005, 15:1277-1288.

15. Bazeley P: Qualitative Data Analysis with NVivo. 2 edition. London: SagePublications; 2007.

16. Ludgrén-Laine H, Salanterä S: Think-aloud technique and protocolanalysis in clinical decision-making research. Qual Health Res 2010,20:565-575.

17. Medical Research Act. [http://www.finlex.fi/fi/laki/kaannokset/1999/en19990488.pdf].

18. Aitken L: Critical care nurses’ use of decision-making strategies. J ClinNurs 2003, 12:476-483.

19. Cooper AB, Joglekar AS, Gibson J, Swota AH, Martin DK: Communication ofbed allocation decisions in a critical care unit and accountability forreasonableness. BMC Health Serv Res 2005, 5:67.

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 9 of 10

20. Reddy MC, Pratt W, Dourish P, Shabot MM: Asking questions: informationneeds in a surgical intensive care unit. Proc AMIA Symp 2002, 647-651.

21. Cardoso LT, Grion CM, Matsuo T, Anami EH, Kauss IA, Seko L,Bonametti AM: Impact of delayed admission to intensive care units onmortality of critically ill patients: a cohort study. Crit Care 2011, 15:R28.

22. Singh M, Nayyar V, Clark P, Kim C: Does after-hours discharge of ICUpatients influence outcome? Crit Care Resusc 2010, 12:156-161.

23. Flaatten H, Hevrøy O: Errors in the intensive care unit (ICU): experienceswith an anonymous registration. Acta Anaesthesiol Scand 1999, 43:614.

24. Flaatten H: Self-reporting of errors and adverse events in the intensivecare unit (ICU). In Organisation and Management of Intensive Care. Editedby: Flaatten H, Putensen C, Moreno RP, Rhodes A. Berlin: MWV MedizinischWissenschaftliche Verlag; 2010:369-373.

25. Ferdinande P: Recommendations on minimal requirements for IntensiveCare Departments. Members of the Task Force of the European Societyof Intensive Care Medicine. Intensive Care Med 1997, 23:226-232.

26. Haupt MT, Bekes CE, Brilli RJ, Carl LC, Gray AW, Jastremski MS, Naylor DF,Rudis M, Spevetz A, Wedel SK, Horst M, Task Force of the American Collegeof Critical Care Medicine, Society of Critical Care Medicine: Guidelines oncritical care services and personnel: recommendations based on asystem of categorization of three levels of care. Crit Care Med 2003,31:2677-2683.

27. Thijs LG: Continuous quality improvement in the ICU: general guidelines.Task Force European Society of Intensive Care Medicine. Intensive CareMed 1997, 23:125-127.

28. Garland A: Improving the ICU: part 1. Chest 2005, 127:2151-2164.29. Rothen HU, Stricker K, Einfalt J, Bauer P, Metnitz PG, Moreno RP, Takala J:

Variability in outcome and resource use in intensive care units. IntensiveCare Med 2007, 33:1329-1336.

30. Qiu Y, Yu P: Nursing information systems: applying usability testing toassess the training needs for nursing students. Methods Inf Med 2007,46:416-419.

31. Johnson C, Zeiger R, Das AK, Goldstein M: Task analysis of writing hospitaladmission orders: evidence of a problem-based approach. AMIA AnnuSymp Proc. 2006, 389-393.

32. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P: TheHawthorne Effect: a randomised, controlled trial. BMC Med Res Methodol2007, 7:30.

33. Westbrook JI, Ampt A: Design, application and testing of the workobservation method by activity timing (WOMBAT) to measure clinicians’patterns of work and communication. Int J Med Inform 2009, 78S:S25-S33.

34. Lauri S, Salanterä S: Developing an instrument to measure and describeclinical decision making in different nursing fields. J Prof Nurs 2002,18:93-100.

doi:10.1186/cc10341Cite this article as: Lundgrén-Laine et al.: Managing daily intensive careactivities: An observational study concerning ad hoc decision making ofcharge nurses and intensivists. Critical Care 2011 15:R188.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit

Lundgrén-Laine et al. Critical Care 2011, 15:R188http://ccforum.com/content/15/4/R188

Page 10 of 10